23rd Annual Student Symposium23rd Annual Student Symposium August 6, 2015August 6, 2015
Harvard Forest Summer Research Harvard Forest Summer Research Harvard Forest Summer Research
Program In EcologyProgram In EcologyProgram In Ecology
23rd Annual Harvard Forest Summer Research Program Symposium
August 6, 2015 Harvard Forest Fisher Museum
Petersham, Massachusetts
Introduction to the Harvard Forest 1
About the Summer Student Research Program 2
2015 Student Seminars and Programs 3
Funding for the Summer Program 4
Symposium Schedule 5
Abstracts 7
Personnel at the Harvard Forest 27
2015 Summer Research Program Students 28
Photographs by Andrew McDevitt, Jenny Hobson
and 2015 Summer Program Participants
1
INTRODUCTION TO THE HARVARD FOREST
Since its establishment in 1907, the Harvard Forest has served as Harvard University’s outdoor classroom and laboratory focused on forest biology and ecology. Through the years, researchers at the Harvard Forest have concentrated on forest management, the development of forest site concepts, the biology of trees, plant ecology, soil processes, forest economics, landscape history, conservation biology, and ecosystem dynamics.
Today, this legacy is continued by faculty, staff, and students who seek to understand historical and modern changes in the forests of New England and beyond. Their research has informed conservation and management as well as enhanced appreciation of forest ecosystems and their histories. This activity is epitomized by the Harvard Forest Long Term Ecological Research (HF LTER) program, which was established in 1988 with funding from the National Science Foundation (NSF).
Physically, the Harvard Forest is comprised of approximately 3,750 acres of land in the north-central Massachusetts town of Petersham and surrounding areas. These acres include mixed hardwood and conifer forests, ponds, streams, extensive spruce and maple swamps, fields, and diverse plantations. Additional land holdings include the 20-acre Pisgah Forest in southwestern New Hampshire (located in the Pisgah State Park), which had been a 300 year-old forest of white pine and hemlock when it was blown down in the 1938 Hurricane; the 100-acre Matthews Plantation in Hamilton, Massachusetts, which is largely comprised of plantations and upland forest; and the 90-acre Tall Timbers Forest in Royalston, Massachusetts.
In Petersham, a complex of buildings that includes Shaler Hall, the Fisher Museum, and the John G. Torrey Laboratories provide office and library space, laboratory and greenhouse facilities, and a lecture room for seminars and conferences. Ten colonial-style houses provide accommodations for staff, visiting researchers, and students. Extensive records, including long-term data sets, historical information, original field notes, maps, photographic collections, and electronic data are maintained in the Harvard Forest Archives.
Administratively, the Harvard Forest is a department of the Faculty of Arts and Sciences (FAS) of Harvard University. Faculty associated with the Forest offer courses through the Department of Organismic and Evolutionary Biology (OEB), the Harvard Kennedy School (HKS), and the Freshman Seminar Program. Close association is also maintained with Harvard University’s Department of Earth and Planetary Sciences (EPS), School of Engineering and Applied Science (SEAS), School of Public Health (SPH), and Graduate School of Design (GSD). The Harvard Forest’s affiliations outside of Harvard University include close ties with the University of Massachusetts departments of Biology, Natural Resource Conservation, and Computer Science; the Marine Biological Laboratory’s Ecosystems Center; and the University of New Hampshire’s Complex Systems Research Center.
The staff and visiting faculty work collaboratively to achieve the research, educational, and management objectives of the Harvard Forest. A management group meets monthly to discuss current activities and to plan future programs. Regular meetings with the HF-LTER science team, weekly research seminars and lab discussions, and an annual ecology symposium provide for an infusion of outside perspectives. The four-member Facilities Crew undertakes forest management and physical plant activities.
2
ABOUT THE 2015 SUMMER RESEARCH PROGRAM IN ECOLOGY
The 2015 Harvard Forest Summer Student Research program, directed by Aaron Ellison, with assistance from Manisha Patel (coordinator) and Andrew McDevitt (proctor), attracted a diverse group of students to receive training in scientific investiga-tions and experience in long-term ecological research. All students worked closely with mentors on various research projects from field and laboratory experiments to computer based software and hardware development. The program included weekly seminars from scientists, a career panel, and a field excursion on land-use history. The summer included an annual trip to the Harvard Museum of Natural History in Cambridge, MA, for a tour of the collections. The Harvard Forest Summer Research Program in Ecology culminates in the Student Research Symposium held on August 6th, 2015, where students present their research findings to an audience of scientists, peers, and family.
2015 Summer Program Students
3
2015 SUMMER PROGRAM SEMINARS AND WORKSHOPS
June: 4 Sessions
Ecological analytics with R — Matt Lau, Harvard Forest
Mon., June 1 Science Check-in — Manisha Patel, Harvard Forest
Thurs., June 4 Forest Walk – David Foster, Harvard Forest
Mon., June 8 Talking Science with the Public (part 1) — Clarisse Hart, Harvard Forest & Marissa Weiss, Science Policy Exchange
Wed., June 10 Talking Science with the Public (part 2) — Clarisse Hart, Harvard Forest & Marissa Weiss, Science Policy Exchange
Wed., June 17 Connecting science and advocacy: Using the decline of whitebark pine as a case study — Diana Tomback, University of Colorado & Harvard Forest Bullard Fellow
Wed., June 24 Patterns in the diversity of epiphytic lichens — Hannah Buckley, Lincoln University (New Zealand) & Harvard Forest Bullard Fellow
Wed., July 1 The PEcAn Project: Putting ecosystem model-data fusion in your pocket — Michael Dietze, Boston University
Wed., July 8 Environmental Stueardship and Sustainable Forestry Practices in Bhutan — Kinley Tshering, Columbia University
Wed., July 15 Scientific Presentation & Poster Workshop — Aaron Ellison, Harvard Forest
Wed., July 22 Nature Sketching — Elizabeth Farnsworth, New England Wild Flower Society
Mon., July 27 Career Panel — Martha Lyman, Natural Resource Conservation Professional & Harvard Forest Bullard Fellow; Robert Clark, Petersham Con-servation Commission and Open Space & Recreation Committee; Tamara Hill-man, National Ecological Observatory Network (NEON)
4
FUNDING FOR THE 2015 SUMMER RESEARCH PROGRAM IN ECOLOGY The Harvard Forest Summer Research Program in Ecology in 2015 was supported by the following organizations: National Science Foundation REU Site: A Forest full of Big Data: the Harvard Forest Summer Research Program in
Ecology 2015-2019 (DBI-1459519) Collaborative Research: Interacting influences of climate, land use, and other disturbances
on regime shifts in forest ecosystems: Holocene dynamics in the northeastern US (DEB-1146207)
Collaborative Research: forecasting and Forestalling Tipping Points in an Aquatic Ecosystem (DEB-1144056)
Collaborative Research: The climate cascade: functional and evolutionary consequences of climatic change on species, trait, and genetic diversity in a temperate ant community (DEB-1136646)
FSML: Walk-up towers for research, education, communication, and outreach at the Harvard Forest (REU Supplement to DBI-1224437)
HFR LTER V: New Science, Synthesis, and Strategic Vision for Society (DEB-1237491) SI2-SSI: Collaborative Research: Bringing End-to-End Provenance to Scientists (ACI-
1450277) Mount Holyoke College Miller Worley Center for the Environment Summer Leadership Fellowship Bryn Mawr College LILAC Summer Internship Fund Harvard University Department of Organismic and Evolutionary Biology G. Peabody “Peabo” Gardner Memorial Fund Faculty of Arts and Sciences
5
THURSDAY, AUGUST 6th FISHER MUSEUM
Aaron Ellison Welcome
9:30 A.M. Session I: Computational Ecology
Marios Dardas
College of the Holy Cross Searching data provenance created in the R scripting environment
Katerina Poulos
Mount Holyoke College Incorporating data analysis workflow into data provenance to increase accessibility
Nathan Justice
Tufts University EcoApps: A tool for exploring ecosystem dynamics, tipping points, and early warning signals
Tempest McCabe
Bryn Mawr College When a tree falls in a forest, do ants care? A look at ant community diversity in response to foundation species loss in two forest systems
~ Break ~
10:45 A.M. Session II: The World of Ants
Roxanne Hoorn
Eckerd College Population density and biodiversity in ant species of the Harvard Forest
Lauren Raiza
Fullerton College Will you be my neighbor? Community distribution of Harvard Forest ants
Megan Brown
Unity College The determinants that limit the northern range of common woodland ants
Jordan Zitnay
Unity College Selective forces that shape physiological tolerance
~ Lunch ~
23rd ANNUAL HARVARD FOREST SUMMER PROGRAM SYMPOSIUM
6
23rd ANNUAL HARVARD FOREST SUMMER PROGRAM SYMPOSIUM
1:00 P.M. Session III: Forests Through Time
Megan Shadley
Mount Holyoke College Long-term history of fire in Southern New England
Olutoyin Demuren
Harvard University Morticulture: Exploring wood after death
Katrina Fernald
Wheaton College (MA) Understory plant community structure and diversity on hurricane tip-up mounds
Mayra Rodriguez- González
University of Puerto Rico — Bayamon
Ecosystem services and hotspot distribution under four different land-use scenarios in Massachusetts, USA
~ Break ~
2:15 P.M. Session IV: Plant Traits
Brittany Cavazos
Rice University Functional trait diversity in an agricultural setting
Harry Stone
Harvard University Trees pushed to their limits: How range and climactic data can be a proxy for functional trait variability and determine species at risk.
Julia Fisher
Bates College Red maple (Acer rubrum) and red oak (Quercus rubra) exhibit spatial and temporal variation in vulnerability of xylem vessels to cavitation
Natalie Gonzalez
Fullerton College Understanding the interactive effects of invasive species and climate change on early Acer rubrum development
~ Break ~
3:45 P.M. Session V: Soil Dynamics
Alana Thurston
Haverford College The effects of soil warming on relative phosphorus availability and mycorrhizal communities
Josia DeChiara
Hampshire College Metatranscriptomic analysis of warmed forest soils suggests divergence of community biogeochemical functions over time
Forrest Lewis
Harvard University Bounding error for total carbon stock estimates in Harvard Forest organic soil
4:30 P.M. Ticks
Aaron Ellison Tick Study
Marissa Saenger Harvard University Isoprene and other biogenic volatile organic compounds: Sources, sinks, and mechanisms for plant uptake and emission at the Harvard Forest
THURSDAY, AUGUST 6th FISHER MUSEUM
7
Megan Brown Unity College
Mentors: Amy Arnett and Aaron Ellison Group Project: How far will they go? Ant foraging distances in New England forests as the
climate warms
The determinants that limit the northern range of common woodland ants
Aphaenogaster picea reaches the northern limits of its distribution in Maine. The causes for the A. picea not found above the 45 degrees N are unknown. Abiotic factors such as climate may limit their ability to move into suitable habitat. In order to determine whether climate influences their range, measures of distribution and abundance were associated with their physiological traits and local environmental conditions. Distribution and abundance were measured by sampling ant colonies in a 50 mile wide transect along 45 degrees N latitude across Maine. In total, 38 plots were sampled, all evenly distributed above and below 45N. 18 A. picea colonies were collected and reared under common garden and subjected to hot and cold treatments to determine their physiological limits. There was no relationship between heat tolerance (CTmax) and any climatic factors such as mean annual temperature, seasonality, and temperature max. For cold tolerance, we predict that lower thermal limits (CTmin) will most likely have a relationship with the distribution and mean annual temperature of A. picea. Climate, mainly temperature, acts as a selective force on cold tolerance but not heat tolerance. Climate may limit the colonization of A. picea north of 45 degrees N latitude.
8
Brittany Cavazos Rice University
Mentors: Dan Flynn and Martha Hoopes Group Project: Functional trait diversity over an agricultural intensification gradient
Functional trait diversity in an agricultural setting
In the last century, New England has become widely reforested. While these forests provide important services, they are encroaching on open areas crucial not only for agricultural use, but also for the conservation of species unique to open habitats. This project sought to determine if the preservation of open land can benefit both conservation and current agricultural practices using plant community data and grazing treatment with dairy cows. The study site was located at the Harvard Farm in Petersham, Massachusetts. Low-intensity grazing, high-intensity grazing, and control treatments were established over three representative vegetation types. In 27 10x10 meter plots, each with three 1x1 meter subplots, community compositions were recorded before and after the grazing treatment using percent coverage. Plant trait data were also collected, including leaf area, height, leaf angle, SLA, leaf dry matter content, and CN ratios. Functional diversity was used to quantify the range of ecosystem services provided within a community. Initial data indicated functional diversity is strongly driven by vegetation type. Over 100 species were identified and species present within a plot ranged from 5 to 35, highly depending on vegetation type. An overall shift in community composition and functional diversity occurred between the primary and secondary census and functional diversity decreased in the grazing treatment. Further monitoring is needed to conclude agricultural practices can be sustainable over a long period of time. Surveying other taxa in the farm is also crucial to understanding how the ecosystem and services provided are changing over time.
9
Marios Dardas College of the Holy Cross
Mentors: Emery Boose and Barbara Learner Group Project: Data Provenance in R
Searching data provenance created in the R scripting environment
Ensuring that data analysis leads to accurate and reproducible results is critical to scientific progress. As scientists explore increasingly sophisticated issues, they often utilize large data sets and complex models for handling the data. The data provenance, the processes performed to achieve reported results, has become increasingly complex, presenting a challenge to establish the authenticity behind scientific results. Scientists can use RDataTracker, an R library to collect provenance from executing R scripts, to create a Data Derivation Graph (DDG), a structured record of the provenance. DDG Explorer, a program that improves accessibility to DDG, assists with assessing the data provenance. Although the DDG displays this information in an organized fashion, it can be difficult to find specific information within the DDG efficiently – especially within DDG containing hundreds or thousands of nodes. To ensure a quick and intuitive experience with finding data, I focused on creating search capabilities for the DDG. The current search extracts information from the DDG and structures it so that information can be queried by the type of data and associated text information. The interface allows users to enter a search which returns a list of search results. Once the user selects a result the node is displayed. Future updates will incorporate searching within the DDG Explorer’s database (implemented with APACHE JENA) by incorporating the SPARQL querying language in addition to an advanced search design to assist with discovering relationships between nodes. Search will assist scientists with debugging programs in addition to quick traversal of the DDG.
10
Josia DeChiara Hampshire College
Mentors: Lauren Alteio, William Rodriguez and Jeffrey Blanchard Group Project: Global warming impacts on soil carbon, nitrogen, and microbiomes
Metatranscriptomic analysis of warmed forest soils suggests divergence of community biogeochemical functions over time
The soil has been shown to be one of the largest carbon stores, of which as much as 50-70% is stored by roots or root-associated microorganisms. As the global temperature continues to rise, the relationship between temperature and microbial carbon cycling is ever pertinent. At the Harvard Forest in Petersham, MA, three experimental warming plots have simulated climate change since 1991, 2003, and 2006 by heating the soil to 5º C above the ambient temperature using underground heating cables. An initial increase in respiration and nitrogen mineralization yielded higher growth rates in plants. As the experiment progressed, plant growth stagnated, hypothetically due a different limiting nutrient. Through measurements of phosphorus content in green and dead leaves, it was observed that phosphorus was being reabsorbed by trees at a higher rate in the heated plots. This led to the hypothesis that phosphorus was stoichiometrically limiting the amount of usable carbon and nitrogen. Phosphorus is found in organisms in phospholipid bilayer, nucleic acids, and adenosine triphosphate, making it an essential biological element. In this study, the role of phosphorus metabolism was examined through analysis of metatranscriptomic data. Nucleic acids were extracted from soils, sequenced, and annotated to determine biological functions. The Mann-Whitney-Wilcoxon test was used to determine genes whose abundance changed with the treatment. These analyses revealed: (1) transcript abundance diverges as heating time increases; (2) changes in gene abundance are evident in in phosphorus and other biogeochemical cycles; and (3) indicator genes for specific metabolic pathways.
11
Olutoyin Demuren Harvard University
Mentors: Audrey Barker Plotkin and Anthony D’Amato Group Project: Long-term forest dynamics following natural or simulated hurricane
disturbance
Morticulture: Exploring wood after death
Coarse woody debris (CWD) serves an important purpose in forest ecosystems by altering resource access by organisms and acting as a carbon pool. The Harvard Forest simulated pulldown experiment’s massive pulse of such wood provides a valuable opportunity to assess CWD sampling methods. Two methods are compared in this study. The line transect method allows rapid measurement of a large area, and involves measuring the diameters of CWD pieces that intersect random sample transects placed in the pulldown and control plots. In the complete census method, diameter and length measurements of individual downed trees are collected from each entire sample plot. Data collected in 2010 and 2015 were used.
CWD sampling methods gave differing volume results, particularly in the pulldown. The line-transect method gave a higher volume than the tree census method in 2010, but a lower volume in 2015. In contrast, the line-transect method provided a consistently lower volume estimate in the control plot. Volume decreased in the pulldown from 2010-2015, regardless of measurement method. The size distributions of the CWD pieces were compared using Kolmogrov-Smirnov tests. The size distribution of the pulldown data varied significantly between the two methods, whereas the diameter distributions in the control did not. Fewer transects are sampled in the pulldown than in the control, and there is greater variation in volume distribution in the pulldown, which may explain these results. In conclusion, both methods can be used to attain a full picture of dead wood in the forest.
12
Understory plant community structure and diversity on hurricane tip-up mounds
Katrina Fernald Wheaton College (MA)
Mentors: Audrey Barker Plotkin and Anthony D’Amato Group Project: Long-term forest dynamics following natural or simulated hurricane
disturbance
Understory plants compose a majority of the vascular plant species in New England hardwood forests. Tree fall gaps and the resources they provide are one major driver in controlling the distribution microenvironments for the understory plant community. Wind events, such as hurricanes, also create tip-up mounds, which alter the topography of the forest floor and create microenvironments for understory plants. This study aims to determine whether the tip-up complexes formed by windthrown trees add to the diversity of understory microenvironments. In particular, it focuses on the species and density of the understory plants on tip-ups. The understory plants shorter than 1 m were sampled on mounds and the adjacent undisturbed forest floor microsite. In preliminary results, the two microsites have a similar number of species, but the average species richness and total plant cover on the mounds is greater than the undisturbed forest floor by p-values of .00011 and 4.6e-7, respectively. These results suggest that mounds provide a more productive and diverse habitat for understory plants. The positive role of tip-up mounds in the forest should be taken into account during post-storm salvage logging operations, which often destroy these important structural changes. Hurricanes are predicted to increase in frequency and severity with climate change. As such, understanding how these events affect our forests can help us prepare anticipate the influence of these events on understory plant community structure and diversity.
13
Julia Fisher Bates College
Mentors: Craig Brodersen and Brett Huggett Group Project: Structure and function of New England forest trees: Predicting future
forest composition by looking back in time
Red maple (Acer rubrum) and red oak (Quercus rubra) exhibit spatial and temporal variation in vulnerability of xylem vessels to cavitation
Vulnerability to cavitation was evaluated along the root to leaf pathway in two common hardwood species, red maple (Acer rubrum) and red oak (Quercus rubra), at the Prospect Hill tract of Harvard Forest in Petersham, Massachusetts, USA. Air seeding pressure analysis was completed on root, trunk, multi-year stem, current-year stem, and petiole tissue of saplings and mature trees of both species. Current year (2015) xylem vessels were measured in all tissue types, and past year (2014) vessels were also measured in multi-year and trunk tissue. There were no differences in air seeding pressures between saplings and mature trees in either species. While Q. rubra exhibited only a slightly higher resistance to cavitation in trunk than in petiole tissue, A. rubrum exhibited higher resistance to cavitation in trunk and multi-year tissue than in petiole and current-year tissue. A. rubrum xylem vessels in current year growth ring were also more resistant to cavitation than vessels from the past year’s growth rings. There was, however, no difference in air seeding pressure between growth rings in Q. rubra. These data suggest that, although diffuse-porous A. rubrum may better adapted to resist cavitation than ring-porous Q. rubra, soft petiole tissue in both species may act as dispensable fuses to cavitation. As the threat of cavitation will increase with the predicted increase in drought frequency and intensity, these data have important implications for future timber management plans and conservation strategies.
14
Natalie Gonzalez Fullerton College
Mentors: Julia Wheeler, Jennifer Albertine, Dustin Haines and Kristina Stinson Group Project: Mediating biological invasion in a global change context
Understanding the interactive effects of invasive species and climate change on early Acer rubrum development
Climate change and non-native plant invasions can strongly impact native plants. However, climate change and species invasion may interact in complex ways, which could affect native species responses. Acer rubrum (red maple) is a common tree species in the northeastern United States where A. petiolata is invading and climate regimes are shifting. For this reason, understanding the possible long-term effects of species invasion and climate change on A. rubrum, particularly in early seedlings, is critical for predicting how recruitment may change over time. Our objectives were to examine how the interaction of A. petiolata invasion and warming impacted germination, survival, phenology, and growth responses of A. rubrum seedlings. We planted A. rubrum in 160 six-inch pots under 4 different treatment types (Control, heated, A. petiolata extract, and heated with A. petiolata) and monitored the seedlings for 8 weeks. Findings for this ongoing experiment suggest that none of the treatments significantly influenced the day of onset of germination. Heating resulted in lower total germination and fewer surviving germinants, but was associated with earlier leaf development and higher leaf counts in later growth. Plants on A. petiolata addition treatments demonstrated higher total germination and more surviving seedlings. Our results suggest that A. petiolata invasion has no detrimental effect on the early development of A. rubrum and heating has later growth benefits. Early stage A. rubrum performance thus may not suffer under mild to moderate A. petiolata invasion, and warming may lead to a longer growing season and increased growth.
15
Roxanne Hoorn Eckerd College
Mentor: Aaron Ellison Group Project: How far will they go? Ant foraging distances in New England forests as the
climate warms
Population density and biodiversity in ant species of the Harvard Forest
Ants play an instrumental part in the forests of New England by aerating the soil and promoting habitat biodiversity. These essential services are not carried out by one genus alone. Rather, it takes a collection of colonies of various species functioning together to sustain a healthy forest. In order to deepen our understanding of the variety and distribution of ant species and their contribution to the local ecosystem, I examined the population density and biodiversity of ant species in the secondary growth deciduous forest of Harvard Forest in Central Massachusetts. To do this, I surveyed ants in five thirty-six square meter plots, and identified all colonies were according to species and habitat type. Colonies were mapped to illustrate their relation to one another within the area of study. The mapping and analysis of these plots showed the presence of four genera and six main species, among which their prevalence varied greatly. Lasius nearticus was found most commonly with a population density of 0.51 colonies of per square meter; Apheanagaster picea was less than half as frequent at a density of 0.23/m2; Lasius umbratus and Myrmica punctiventris were both present at a density of 0.4/m2; with very few occurrences of Lasius alienus at 0.02/m2 and Formica neogagates at 0.01/m2. These findings suggest that there is a combination of factors specific to individual specie preferences, including habitat type and availability, which shape the population density and biodiversity of ants present on the forest floor.
16
Nathan Justice Tufts University
Mentors: Matthew Lau and Aaron Ellison Project: Modeling tipping points and regime shifts in ecological systems
EcoApps: A tool for exploring ecosystem dynamics, tipping points, and early warning signals
To facilitate the investigation of ecosystem tipping points across the scientific community, this project's goal is to lead the development of a computational tool that will allow researchers, regardless of their familiarity with formal programming languages, the ability to simulate ecosystem dynamics and perform tipping point and early warning signal analyses. Ostensibly stable ecosystems are sometimes susceptible to abrupt and drastic changes from one state to another. These rapid state changes exhibit tipping points (sometimes referred to as breakpoints or changepoints), which represent a threshold for a change in the dynamics demonstrated by the ecosystem's state variables. The complexity of ecosystem state changes is compounded by the knowledge that alternative states are often not transient. Generally, humans and other wildlife are accustomed to a preferable ecosystem state, and tipping points do not provide sufficient opportunity for adaptation. Unexpected tipping points are increasingly more likely to occur as human impacts continue to affect and alter ecosystem dynamics across the biosphere. Mounting ecological research suggests tipping points can be simulated with mathematical and statistical models. In addition, simulations of these models have the potential to illuminate early warning signals, which are probable indicators that proceed an imminent tipping point. The product of this project, EcoApps, is a series of web applications implemented using the Shiny framework in R. Each application simulates a distinct ecosystem model. Users are able to explore the model by manipulating parameters and perform tipping point an early warning signal analyses on the simulation's components (observed data, trend, periodicity, and residuals).
17
Forest Lewis Harvard University
Mentors: Evan Goldman and J. William Munger Project: Dynamics of forest carbon on annual to decadal times
Bounding error for total carbon stock estimates in Harvard Forest organic soil
The terrestrial biosphere serves a fundamental role in the global carbon cycle by releasing and absorbing greenhouse gases. At the Harvard Forest, a deciduous forest in Central Massachusetts, biometric measurements have quantified above-ground organic carbon mass, but soil carbon has been less studied. In June-July 2015, I randomly selected 7 of 34 biometric plots from the footprint of the Environmental Measurement Station (EMS) Tower based on tree species variability to sample for soil organic matter (SOM). The cores, which were paired with samples from the 2014 summer survey to minimize spatial variability, were dried, sieved, root picked, ground, and ashed to determine SOM. The organic layer is defined as greater than 40% SOM and is traditionally isolated by separating cores at the point of color and textural change that indicates the mineral layer. Due to difficulty in delineating the organic and mineral layers, previous studies appear to underestimate the total soil carbon stock at the Harvard Forest. The 2014 summer survey estimated 37.36 MgC/ha in the organic layer, but my collocated measurements found 39.752 ± 1.051 MgC/ha. Using updated methods that allow for a more precise delineation as well as upper and lower bounds, changes over 1.051 MgC/ha can be detected on yearly and decadal time scales. By repeating this more precise way of constraining total carbon values, we can ascertain interannual variability in the organic layer carbon stock, which could shed light into the global carbon budget.
18
Tempest McCabe Bryn Mawr College
Mentors: Sydne Record and Aaron Ellison Project: Foundation tree species affects on arthropod communities
When a tree falls in a forest, do ants care? A look at ant community diversity in response to foundation species loss in two forest systems
Eastern hemlock and the oak genus are foundational trees that are under threat from hemlock woolly adelgid (HWA) and sudden oak death, respectively. This experiment looks at the response of ant community composition to simulated impacts of pests and pathogens in the Harvard Forest Hemlock Removal Experiment (HF-HeRE) in Massachusetts and the Future of Oak Forests experiment at Black Rock Forest (BRF) in New York. The HF-HeRE consists of hardwood and hemlock controls, simulated adelgid infestation (hemlocks girdled), and salvage logging (removal of hemlock and merchantable hardwoods) plots. At BRF four treatments investigate effects of oak loss: all oaks girdled, 50% of oaks girdled, all non-oaks girdled, and a control (oaks and non-oaks were unmanipulated). At HF ants were sampled from 25 pitfall traps every year from 2003-2015, and at BRF ants were hand sampled during five years between 2006-2015. A non-parametric multivariate Analysis of Variance (npMANOVA) found significant effects of year, treatment, plot nested within treatment, and treatment year×treatment interaction. Non-metric multidimensional scaling (NMDS) plots of HF-HeRE treatments over time suggest that girdled and logged plots became more similar to hardwood control three years post-treatment and that hemlock control plots became more similar to hardwood control plots 6-8 years post-treatment and after the infestation of HWA to the forest In 2010. In contrast, an npMANOVA for BRF did not reveal any significant effects of treatment, year, or their interaction. These results suggest ant communities may be more resilient to the loss of oaks than to the loss of hemlocks.
19
Katerina Poulos Mount Holyoke College
Mentors: Emery Boose and Barbara Lerner Group Project: Data Provenance in R
Incorporating data analysis workflow into data provenance to increase accessibility
Scientific progress requires the validation, authentication, and reproduction of results to uphold the quality of scientific research. One technique to make large datasets and complex data analyses more intelligible is to capture the data provenance, or the history of a digital object. The current tools developed by the Harvard Forest team to capture and analyze provenance are RDataTracker and DDGExplorer, respectively. These tools create and use a Data Derivation Graph (DDG), a metadata structure that captures data provenance. However, these DDGs may be difficult to comprehend for many R scripts as comprehensibility of a DDG may be inversely proportional to the size and complexity of the script. The objective of this project is to make provenance shown by the DDG more intelligible by breaking it down into multiple sections that reflect typical data analysis workflow. This automated system takes advantage of a common workflow of an R script developed by analyzing various R scripts from the Harvard Forest community. The current interpretation of the typical data analysis workflow was identified and consists of: initialization, data preparation, data analysis, plotting, and output. Building upon existing tools from RDataTracker, my program will provide a higher abstraction to R scripts with little to no user intervention by labelling the portions of the DDG that correspond to these different data analysis activities. By compartmentalizing provenance into comprehensible bits using lingo known by all scientists, a potentially large and complex DDG will be collapsed into more readily coherent sections which will increase accessibility.
20
Lauren Raiza Fullerton College
Mentor: Aaron Ellison Group Project: How far will they go? Ant foraging distance in New England forests as the
climate warms
Will you be my neighbor? Community distribution of Harvard Forest ants
Ants are complex social creatures that live in a variety of habitats within the Harvard Forest in Central Massachusetts. They occupy trees, soil, and leaf litter and are responsible for breaking down decaying matter, aerating and producing soil, and even dispersing seeds. With all of these jobs to do and places to be, it is no wonder that the different genera and species of ants that occupy the Forest live very different lives. These ants’ lives intertwine when competition for habitat or other resources come into play, and these interactions may affect their selection of nesting sites. To examine which species of ants were likely to nest near one another, five thirty-six square meter plots were sampled within the forest to examine spatial distribution of nest sites. By examining spatial relationships between Aphaenogaster picea, Formica neogagates, Lasius nearcticus, Lasius umbratus, and Myrmica punctiventris much can be learned about the behavior of these ants and how their distribution can affect forest dynamics. Further investigation of these interspecies relationships may provide insight into nest site and neighbor selection behaviors in relation to competition.
21
Mayra Rodriquez-Gonzalez University of Puerto Rico—Bayamon
Mentors: Luca Morreale, Meghan Blumstein, Matthew Duveneck and Jonathan Thompson Group Project: Land use effects on New England forest composition and structure
Ecosystem services and hotspot distribution under four different land-use scenarios in Massachusetts, USA
Simulating land-use effects on ecosystem services can inform land management and policy decisions. Using four land-use scenarios for the state of Massachusetts, this study maps land-use impacts on the state’s capacity to provision ecosystem services over time, and emphasizes changes in the abundance and distribution of service “hotspots,” which are areas that provide a high value for multiple services. The scenarios depict alternative trajectories of land-use over the next 50 years, including different rates, intensities, and spatial arrangements of development, harvest, agriculture and conservation. Throughout the simulations, the Recent Trends scenario had the least area of hotspots, but those areas were comparatively stable. In contrast, the Opportunistic Growth scenario had the greatest loss of high-value areas, while the Regional Self-Reliance scenario had the greatest increase in high-value areas. In all scenarios, the area of hotspots increased through time, which reflects the concentration of services on to less area and, in some cases, the expansion of timber and agriculture. We conclude that, in aggregate, land-use decisions produce distinct patterns of ecosystem service provisioning. However, we also note that the estimated effects of land-use on services are strongly affected by the modeling approach. Nonetheless, this type of multi-service, multi-scenario assessment can inform policymakers regarding the potential consequences of land-use and assist them in identifying priority conservation zones.
22
Marissa Saenger Harvard University
Mentor: Karena McKinney Project: Exchange of biogenic volatile organic compounds between the forest and the
atmosphere
Isoprene and other biogenic volatile organic compounds: Sources, sinks, and mechanisms for plant uptake and emission at the Harvard Forest
All plants emit biogenic volatile organic compounds (BVOCs), hydrocarbons that interact with atmospheric compounds to form products such as ozone and aerosol particles. Isoprene comprises roughly half of BVOC emissions worldwide, and its reactions affect tropospheric ozone levels, which influences climate and air quality. BVOCs are emitted in varying levels by different plant species: grapes (vinis vinifera) are known to lack isoprene emissions, while red maple (acer rubrum) and red oak (quercus rubra) exhibit low and high isoprene emission levels, respectively. The goals of this project were to determine isoprene and other BVOC emission levels among these species along with photosynthesis and transpiration rates and to determine methods of emission and uptake. The experiments used a protontransferreaction mass spectrometer (PTRMS) to detect levels of isoprene and other BVOCs (including acetone, monoterpenes, acetaldehyde, and oxygenated terpenes), sampled from bag chambers enclosing plant leaves. Each species was tested individually for 24 hours, repeated several times using different leaves or specimens. CO2 and water concentrations, measured simultaneously using a Li840A instrument, were used to calculate transpiration and photosynthesis rates of the enclosed leaves and test for correlations between these processes and BVOC emissions. Stomatal conductance was calculated from transpiration rates and used to assess the role of stomata on BVOC emissions. Findings were consistent with expectations regarding relative isoprene emission levels among the three species. The data collected indicated peaks in most BVOC emissions during daylight hours, along with transpiration and photosynthesis rates, indicating correlation between BVOC emissions and leaf transpiration and photosynthesis.
23
Megan Shadley Mount Holyoke College Mentor: Wyatt Oswald
Project: Using paleoecology to understand long-term forest dynamics in New England
Long-term history of fire in Southern New England
For decades, some ecologists and conservationists have asserted that Native Americans commonly utilized prescribed fire as a method of forest management in New England long before European settlement (Day 1953, Abrams 1992). Accordingly, fire would have been prevalent during peaks in human population in the middle and late Holocene (Munoz et al. 2010), when large numbers of people purposefully burned forests to favor food-producing plants and to facilitate hunting and travel. Alternatively, the effects of changing vegetation and/or climate may have controlled fire by impacting fuel conditions. To date, few lake-sediment charcoal records have been available to test these hypotheses. The influx of charcoal pieces provides a proxy for the relative abundance of fire events, particularly in a localized watershed area. In this study we analyzed charcoal preserved in the sediments of three Massachusetts ponds: Green Pond in Montague, and Jernagen’s and Lily ponds on Martha’s Vineyard. The two sites on Martha’s Vineyard would have had particularly high indigenous populations during the late Holocene. We found that all three sites had low charcoal influx values during intervals of high human populations, suggesting that the human control of fire was less important than previously believed. Continued paleoenvironmental studies of cores from Ware Pond in Marblehead, MA, may yield further insights into the role of climate change in long-term variations in fire activity.
24
Harry Stone Harvard University
Mentors: Dan Flynn and Elizabeth Wolkovich Project: Tree functional traits across climate gradients
Trees pushed to their limits: How range and climactic data can be a proxy for functional trait variability and determine species at risk
Broadleaf temperate forests in the Northeast are an important economic and ecological zone facing an increasing risk due to intensifications in temperature and a shifting climate. While the future impact is not well understood, a latitudinal gradient in the biome could be used to predict temporal changes and how the forest and specific species may respond. The intended goal of this research is to determine how woody deciduous species react differently to being pushed to their climatic limits. Morphological, physiological, and chemical traits were measured of 393 individuals stemming from 38 species. Sampling occurred at the Harvard Forest and the University of Montréal Laurentides field station to create a 3.70 degree latitudinal shift to determine how a harsher climate impacts viability. GIS tools were used to compare the two field sites to the range extents of the studied species. Spatial and temperature-linked distance to end of range predicted functional trait changes and variability on a species to species basis. Trees and shrubs with a larger range were predicted to have more phenotypic plasticity and thus resistant to change across the latitudinal range. Species with a greater northern extent, such as Betula papyrifera and Populus tremuloides, are buoyed more against the climate shift and their functional trait response is more muted. Results show a decrease in specific leaf area, stem specific density, leaf dry matter content, and percent nitrogen between the Harvard Forest and Quebec highlighting a mainly conservative growth strategy.
25
Alana Thurston Haverford College
Mentors: Michael Bernard, William Werner and Jerry Melillo Group Project: Global warming impacts on soil carbon, nitrogen, and microbiomes
The effects of soil warming on relative phosphorus availability and mycorrhizal communities
Soil warming, as an effect of climate change, has the potential to alter soil processes, nutrient availability, and soil community composition. The Barre Woods Soil Warming Experiment was designed to monitor these effects, and in the first decade of the experiment, nitrogen availability increased, leading to accelerated aboveground tree growth. However, in the past two years, woody increment has returned to control rates and phosphorus resorption in leaf tissue has increased, leading us to hypothesize that chronic soil warming has led to progressive phosphorus limitation. We developed two hypotheses that could explain phosphorus limitation – reduced phosphate bioavailability or shifts in mycorrhizal exploration types. We used ion exchange membranes to assess the effect of soil warming on phosphate availability, and found that there were no significant differences between treatments. However, by characterizing 59 ectomycorrhizal morphotypes, we found that there was a significant shift to contact exploration species in the heated plot over both short and long-distance. This indicates that trees in the heated plot might not have as much access to phosphate because they aren’t connected to a large hyphal network. Alternatively, the shift in mycorrhizal species could change the enzymes produced in the soil, which would affect the breakdown of organic matter and nutrient availability. We tested the potential enzyme activity of phosphatase – an enzyme produced by fungi to make phosphate ions available – but found that there was no significant difference between treatments.
26
Jordan Zitnay Unity College
Mentors: Amy Arnett and Aaron Ellison Group Project: How far will they go? Ant foraging distances in New England forests as
the climate warms
Selective forces that shape physiological tolerance
Ants occupy a wide variety of environments but their abundance and distribution can be limited by numerous biotic and abiotic factors. In particular, for the common woodland ant (Aphaenogaster picea), the northern edge of its range does not extend past 45°N in Maine despite suitable nesting habitat. Limitations in physiological tolerance for temperature extremes may limit their range expansion. To determine whether climatic variables shape thermal tolerance, we associated measures of hot and cold temperature tolerances for 18 colonies that were collected along the species’ northern range under common garden conditions with climatic factors. High temperature tolerance was measured as the loss of motor function during slow ramping temperature treatment (CTmax) and cold temperature tolerance was measured as recovery time after exposure to cold shock (CTmin). There was no relationship between CTmax and mean annual temperatures for A. picea colonies collected at their northern range boundary. However, there was a negative relationship between cold tolerance of ants and mean annual temperature. Temperature acts as a force of selection for cold but not heat tolerance on the populations of a common woodland ant at the northern end of their range.
27
Jay Aylward Audrey Barker-Plotkin Daniel Bishop Emery Boose Jeannette Bowlen Laurie Chiasson Betsy Colburn Elaine Doughty Matthew Duveneck Edythe Ellin Aaron Ellison Kathy Fallon Lambert David Foster Lucas Griffith Brian Hall Julie Hall Clarisse Hart Jenny Hobson David Kittredge Oscar Lacwasan Matthew Lau Ronald May Roland Meunier Alisha Morin Luca Morreale John O’Keefe David Orwig Julie Pallant Manisha Patel Neil Pederson Joshua Plisinski Lisa Richardson Pep Serra Diaz Pamela Snow Jonathan Thompson Greta VanScoy Mark VanScoy Marissa Weiss John Wisnewski Peter Yesmentes Tim Zima Harvard University Affiliates Richard Forman Noel Michele Holbrook William Munger Andrew Richardson Steven Wofsy Elizabeth Wolkovich
Research Assistant Site and Research Manager Research Assistant Information Manager Accountant Administrative Assistant Aquatic Ecologist Research Assistant Post-Doc Director of Administration Senior Ecologist/Community Ecologist Director of Science and Policy Exchange Director of Harvard Forest Woods Crew GIS Specialist Assistant Data Manager Outreach and Development Manager Secretary Forest Policy Analyst Woods Crew Post-Doc Woods Crew Woods Crew Accounting Assistant Research Assistant Museum Coordinator (Emeritus) Senior Ecologist/Forest Ecologist Information Technology and Archives Administrator Lab Manager and Summer Program Coordinator Senior Ecologist/Forest Ecologist Research Technician Accounting Assistant Post-Doc Schoolyard Program Coordinator Senior Ecologist/Landscape Ecologist Museum Coordinator Field Instrument Specialist Science Policy Exchange Coordinator Woods Crew Supervisor Summer Program Assistant Cook Summer Program Cook Graduate School of Design Organismic and Evolutionary Biology School of Engineering and Applied Sciences Organismic and Evolutionary Biology School of Engineering and Applied Sciences OEB and Arnold Arboretum
PERSONNEL AT THE HARVARD FOREST - 2015
28
Megan Brown Unity College
Brittany Cavazos Rice University
Marios Dardas College of the Holy Cross
Josia DeChiara Hampshire College
Olutoyin “Olu” Demuren Harvard University
Katrina “Katie” Fernald Wheaton College (MA)
Julia Fisher Bates College
Natalie Gonzalez Fullerton College
Roxanne Hoorn Eckerd College
2015 SUMMER RESEARCH PROGRAM STUDENTS
29
2015 SUMMER RESEARCH PROGRAM STUDENTS
Simone Johnson Lincoln University
Emily Keenan University of Michigan
Bruce McAlister Univ. of Massachusetts Amherst
Sofie McComb University of Texas at Austin
Alison Ochs Mount Holyoke College
Maria Orbay-Cerrato Brown University
Luis Perez Harvard College
Ariel da Cruz Reis Saint Michael’s College
Harry Stone Harvard University
Nathan Justice Tufts University
Forrest Lewis Harvard University
Tempest “Tess” McCabe Bryn Mawr College
Katerina “Lia” Poulos Mount Holyoke College
Lauren “Cody” Raiza Fullerton College
Mayra Rodriquez-Gonzalez Univ. of Puerto Rico—Bayamon
Marissa Saenger Harvard University
Megan Shadley Mount Holyoke College
30
2015 SUMMER RESEARCH PROGRAM STUDENTS
Alana Thurston Haverford College
Jordan Zitnay Unity College
2015 SUMMER RESEARCH PROGRAM PROCTOR
Andrew McDevitt Illinois State University
